mjsalnic-Data Replication Assignment

---

Development of hatching ability in red-eyed treefrogs: escape from complications

---

Authors: MJ Salazar-Nicholls^1,3, KD Escobar² & KM Warkentin³

Affiliations: 1. Pontificia Universidad Católica, Ecuador; 2. Western Connecticut State University; 3. Boston University – majonicholls2909@gmail.com – Warkentin Lab

Brief Introduction

Hatching early allows embryos to escape threats to eggs, but increases risks to larvae. Red-eyed treefrogs hatch by rapidly releasing enzymes to digest a small hole in their membrane, then squeezing out aided by turgor pressure. Displacement from the initial hole can occur spontaneously and in predator attacks, complicating hatching by capsular collapse as fluid escapes. To assess developmental changes in ability to recover from such complications, we manually displaced 3-5 day old embryos during hatching, interrupting exit through their initial hole, and recorded macro-video. To induce hatching of younger, vibration-insensitive embryos, we submerged individual eggs in hypoxic water. For older embryos, we used a blunt probe to jiggle eggs within clutches. Once an embryo began hatching, we manually displaced it, moving its snout away from the forming or just-formed hole. We analyzed videos to assess if and how embryos escaped. The least-developed embryos rarely escaped from collapsed capsules but performance improved rapidly; late stages always escaped. Embryos escaped by either re-finding their first hole or making a second. The post-displacement delay to start an exit through the first hole was similar across stages. Even at the earliest stages, a few individuals made second holes, but the likelihood of doing so increased rapidly, to about 75%. More developed embryos spent less effort searching for their first hole; they made second holes sooner and more efficiently. They also passed through the membrane more rapidly, with less body compression in transit. Along with high risk after hatching, poor ability to escape from hatching complications may select against unnecessary hatching attempts at early developmental stages.

wasp attacking Agalychnis callidryas clutch

Objective

Assess developmental changes in ability to recover from hatching complications in Red-eyed treefrogs embryos

Methods

• Collected young A. callidryas clutches from pond, maintained in lab at Smithsonian Tropical Research Institute, Gamboa, Panama
• Induced individual 3– 5 day old embryos to begin hatching, using hypoxia and mechanosensory stimuli
• Manually displaced embryo in egg, interrupting exit through initial rupture
• Recorded macro-video until embryo exited its capsule; rescued individuals that failed to hatch after 30 minutes
• Video analysis: We analyzed recordings in QuickTime Player (Apple Inc), event timing was quantified from frame numbers. Quantified events were first membrane rupture time, position changes after displacement, second rupture time, body compression observed during exit, compression duration time, exit duration as defined in Cohen et al (2016).

Flooding	Jiggling
we submerged younger eggs in hypoxic water (N = 25, 27 at ages 3 and 4 d)	we used a blunt probe to move older eggs on clutches (N = 64, 51 at ages 4 and 5 d)

R1: rupture one

R2: rupture two

Outline

This is an analysis of the data used for a poster presentation at SICB 2017, which now is part of a manuscript. The statistical analyses presented in the poster were made using JMP-data analysis sorfware The statistical analysis in made for SICB includes: boxplots, contingency Analysis (Mosaic plots and contingency tables), Wilcoxon/Kruskal-Wallis (Rank Sum) test, Fisher´s exact Test. I will replicate all of the analyses presented in the original poster.

1. Load and data preparation
2. Exporatory analysis
3. Results

• 3.1. Escape-hatching success after complications: compare the hatching success of embryos after displacement across different developmental stages

     * Mosaic plots, contingency analysis
  
     * Pearson's Chi-squared test

• 3.2. Multiple membrane ruptures: analyzed the ability to make more membrane ruptures after displacement across different developmental stages, comparison between treatments

     * Mosaic plots, contingency analysis
     * Wilcoxon test

• 3.3. Position changes after displacement – search for initial rupture site: compare the behavior (body positions changes) after the embryo lost the first rupture. How many times the embryos -across different developmental stages- move around the egg before making another hole or until the time when need to be rescued

     * Box plots
     * Wilcoxon test

• 3.4. Body compression during exit: compare body compression observation and duration across different developmental stages

     * Mosaic plots, contingency analysis
     * Box plots
     * Wilcoxon test

• 3.5. Exit duration: compare the speed of the hatching process (after displacement) across different developmental stages

     * Box plots
     * Wilcoxon test

4. Figure
5. Conclusions
6. References

Packages required

• curl
• dplyr
• ggplot2
• plotly
• EnvStats
• MASS
• coin
• vcd
• vcdExtra

1. Load and data preparation

1.1 Data loading

f <- curl("https://raw.githubusercontent.com/mjsalnic/mjsalnic-data-replication-assignment/master/HC2016R.csv")
HC2016R <- read.csv(f, header = TRUE, sep = ";",dec=",", stringsAsFactors = FALSE)
kable(head(HC2016R))

Treatment	ClutchN	Age	Stage	comst	groupst	DecimaClocklHours	AccumshakingtimeRup1Fra	AccumShakingR1sec	NumPosChanADBefore2ndHole	AccumADshakingR	AccumshakingtimeRup2Fra	AccumShakingR2sec	DisEndRup2Fr	DisEndHatchingStartFr	CompDurFr	CompressionSec	CompressionSeen	EXITDURATIONFr	EXIT.DURATION.StarttoFullyOutsec	NumHoles	FirstSecondHoleExit	Rescue	X	X.1
Flooding	462	3	2	2-3	2-3	20.23	4	0.13	3	NA	NA	NA	NA	1407	NA	NA	0	2442	81.4000000	1	First	0
Flooding	470	3	3	2-3	2-3	21.02	290	9.67	7	NA	0	0	NA	4838	NA	NA	0	1318	43.9333333	1	First	0
Flooding	386	4	4	4-5	4-7	0.15	30	1.00	1	NA	0	0	NA	916	6	0.20	1	342	11.4000000	1	First	0
Flooding	397	4	5	4-5	4-7	0.82	395	13.17	1	NA	NA	NA	NA	5182	NA	NA	0	4	0.1333333	1	First	0
Flooding	102	4	5	4-5	4-7	10.85	105	3.50	17	NA	0	0	NA	6511	7	0.23	1	3995	133.1666667	1	First	0
Flooding	187	4	4	4-5	4-7	16.12	260	8.67	1	NA	NA	NA	NA	2222	NA	NA	0	157	5.2333333	1	First	0

1.2 Preparing the data frame to analysis

• Transforming numerical variables to factorial

HC2016R$Stage<- factor(HC2016R$Stage)
HC2016R$Age<- factor(HC2016R$Age)
HC2016R$NumHoles<- factor(HC2016R$NumHoles)
HC2016R$Rescue<- factor(HC2016R$Rescue)

str(HC2016R)

## 'data.frame':    167 obs. of  25 variables:
##  $ Treatment                       : chr  "Flooding" "Flooding" "Flooding" "Flooding" ...
##  $ ClutchN                         : int  462 470 386 397 102 187 272 268 239 244 ...
##  $ Age                             : Factor w/ 3 levels "3","4","5": 1 1 2 2 2 2 2 2 2 2 ...
##  $ Stage                           : Factor w/ 6 levels "2","3","4","5",..: 1 2 3 4 4 3 4 3 5 5 ...
##  $ comst                           : chr  "2-3" "2-3" "4-5" "4-5" ...
##  $ groupst                         : chr  "2-3" "2-3" "4-7" "4-7" ...
##  $ DecimaClocklHours               : num  20.23 21.02 0.15 0.82 10.85 ...
##  $ AccumshakingtimeRup1Fra         : int  4 290 30 395 105 260 285 0 307 170 ...
##  $ AccumShakingR1sec               : num  0.13 9.67 1 13.17 3.5 ...
##  $ NumPosChanADBefore2ndHole       : int  3 7 1 1 17 1 1 14 16 1 ...
##  $ AccumADshakingR                 : int  NA NA NA NA NA NA NA NA NA NA ...
##  $ AccumshakingtimeRup2Fra         : int  NA 0 0 NA 0 NA NA NA 385 NA ...
##  $ AccumShakingR2sec               : num  NA 0 0 NA 0 ...
##  $ DisEndRup2Fr                    : int  NA NA NA NA NA NA NA NA NA NA ...
##  $ DisEndHatchingStartFr           : int  1407 4838 916 5182 6511 2222 213 8227 5455 187 ...
##  $ CompDurFr                       : int  NA NA 6 NA 7 NA NA NA NA NA ...
##  $ CompressionSec                  : num  NA NA 0.2 NA 0.23 NA NA NA NA NA ...
##  $ CompressionSeen                 : int  0 0 1 0 1 0 0 0 0 0 ...
##  $ EXITDURATIONFr                  : int  2442 1318 342 4 3995 157 96 NA NA 147 ...
##  $ EXIT.DURATION.StarttoFullyOutsec: num  81.4 43.933 11.4 0.133 133.167 ...
##  $ NumHoles                        : Factor w/ 2 levels "1","2": 1 1 1 1 1 1 1 1 2 1 ...
##  $ FirstSecondHoleExit             : chr  "First" "First" "First" "First" ...
##  $ Rescue                          : Factor w/ 2 levels "0","1": 1 1 1 1 1 1 1 1 1 1 ...
##  $ X                               : chr  "" "" "" "" ...
##  $ X.1                             : chr  "" "" "" "" ...

• Subsetting data frame

let’s subset the data into 2 outcome groups: success_hat (hatched embryos after displacement) and failed_hat(embryos unable to hatch after displacement)

success_hat<-subset(HC2016R, Rescue==0, na.rm=T)
failed_hat<-subset(HC2016R, Rescue==1, na.rm=T)

let’s subset the data into 2 treatment groups: flooding and jiggling

flooding<-subset (HC2016R, Treatment=="Flooding", na.rm=T)
jiggling<-subset(HC2016R, Treatment=="Jiggling", na.rm=T)

2. Exploratory analysis

Compare number of holes by Age

mosaic(~ NumHoles + Age, data = HC2016R, main = "Escape-hatching success after complications", shade= TRUE,legend=TRUE)

Compare number of holes by Stage

mosaic(~ NumHoles + Stage, data = HC2016R, main = "Escape-hatching success after complications", shade= TRUE,legend=TRUE)

Comparing position changes- Should we use parametric or non-parametric test?

par(mfrow=c(3,2))
hist(HC2016R$NumPosChanADBefore2ndHole[HC2016R$Stage=="2"])
hist(HC2016R$NumPosChanADBefore2ndHole[HC2016R$Stage=="3"])
hist(HC2016R$NumPosChanADBefore2ndHole[HC2016R$Stage=="4"])
hist(HC2016R$NumPosChanADBefore2ndHole[HC2016R$Stage=="5"])
hist(HC2016R$NumPosChanADBefore2ndHole[HC2016R$Stage=="6"])
hist(HC2016R$NumPosChanADBefore2ndHole[HC2016R$Stage=="7"])

Non-parametric

Exit duration in seconds- Should we use parametric or non-parametric test?

par(mfrow=c(3,2))
hist(success_hat$EXIT.DURATION.StarttoFullyOutsec[success_hat$Stage=="2"])
hist(success_hat$EXIT.DURATION.StarttoFullyOutsec[success_hat$Stage=="3"])
hist(success_hat$EXIT.DURATION.StarttoFullyOutsec[success_hat$Stage=="4"])
hist(success_hat$EXIT.DURATION.StarttoFullyOutsec[success_hat$Stage=="5"])
hist(success_hat$EXIT.DURATION.StarttoFullyOutsec[success_hat$Stage=="6"])
hist(success_hat$EXIT.DURATION.StarttoFullyOutsec[success_hat$Stage=="7"])

Non-parametric

3. Results

3.1 Escape-hatching success after complications

Original Data:

3.1.1 Mosaic plot - graphical representation of the frequency of successul hatching and unable to hatch after displacement experiment (combined experiments)

mosaic(~ Rescue + Stage, data = HC2016R, main = "Escape-hatching success after complications", shade= TRUE,legend=TRUE)

totals=table(HC2016R$Stage,HC2016R$Rescue)
mosaicplot(totals, main = "Escape-hatching success after complications", xlab = "Embryonic developmental stage", ylab = "proportion of embryos", col=c("light green", "black"))

#where 0 means no rescue needed, embryo able to hatch by itself and 1 means rescue after displacement experiment 

  * 1= Failed to hatch 

  * 0= Succesfully hatched embryos 

  * Original JMP: P < 0.0001
  * Replication: P < 0.0001

2. Contingency table- Escape-hatching success after complications

tab1<-table(HC2016R$Stage, HC2016R$Rescue)
prop.table(tab1)*100

##    
##             0         1
##   2  2.395210  4.191617
##   3  5.988024  1.796407
##   4 11.377246  0.000000
##   5 20.359281  1.796407
##   6 20.958084  0.000000
##   7 31.137725  0.000000

chisq.test(tab1)

## Warning in chisq.test(tab1): Chi-squared approximation may be incorrect

## 
##  Pearson's Chi-squared test
## 
## data:  tab1
## X-squared = 60.99, df = 5, p-value = 7.588e-12

fisher.test(tab1)

## 
##  Fisher's Exact Test for Count Data
## 
## data:  tab1
## p-value = 9.464e-09
## alternative hypothesis: two.sided

3.2 Multiple membrane ruptures

Original Data

1. Treatments together

tab<-table(HC2016R$Stage, HC2016R$NumHoles)
prop.table(tab)*100

##    
##             1         2
##   2  4.790419  1.796407
##   3  1.796407  5.988024
##   4  2.395210  8.982036
##   5  7.185629 14.970060
##   6  8.982036 11.976048
##   7  7.784431 23.353293

chisq.test(tab)

## Warning in chisq.test(tab): Chi-squared approximation may be incorrect

## 
##  Pearson's Chi-squared test
## 
## data:  tab
## X-squared = 12.719, df = 5, p-value = 0.02616

fisher.test(tab)

## 
##  Fisher's Exact Test for Count Data
## 
## data:  tab
## p-value = 0.03435
## alternative hypothesis: two.sided

By treatment

2. Flooding treatment- Mosaic plot- multiple membrane ruptures

totals=table(flooding$Stage,flooding$NumHoles)
mosaicplot(totals, main = "Multiple membrane ruptures", xlab = "Embryonic developmental stage", ylab = "Proportion of embryos", col=c("light green", "black"))

Compare number of holes by stage- Flooding treatment

mosaic(~ NumHoles + Stage, data = flooding, main = "Escape-hatching success after complications", shade= TRUE,legend=TRUE)

  * Original JMP: p-value = 0.011
  * Replication: p-value = 0.018
  
  

2.1 Flooding treatment- Contigency table

tab2<-table(flooding$Stage, flooding$NumHoles)
tab2

##    
##      1  2
##   2  8  3
##   3  3 10
##   4  2 14
##   5  2  9
##   6  0  1
##   7  0  0

3. Jiggling treatment- Mosaic plot

totals=table(jiggling$Stage,jiggling$NumHoles)
mosaicplot(totals, main = "Multiple membrane ruptures", xlab = "Embryonic developmental stage", ylab = "Proportion of embryos", col=c("light green", "black"))

3.1 Jiggling treatment- Contigency table

tab3<-table(jiggling$Stage, jiggling$NumHoles)
fisher.test(tab3)

## 
##  Fisher's Exact Test for Count Data
## 
## data:  tab3
## p-value = 0.1508
## alternative hypothesis: two.sided

summary(tab3)

## Number of cases in table: 115 
## Number of factors: 2 
## Test for independence of all factors:
##  Chisq = NaN, df = 5, p-value = NA
##  Chi-squared approximation may be incorrect

  * Original JMP: p-value= 0.16
  * Replication: p-value = 0.1508
  
  

3.3 Position changes after displacement – search for initial rupture site

Original Data

Outcome (S2-3)

par(mfrow=c(1,2))
boxplot(NumPosChanADBefore2ndHole~Stage,success_hat, main ="Successfully hatched" , ylab = "Position changes before 2nd hole")

boxplot(NumPosChanADBefore2ndHole~Stage,failed_hat, main="Rescued embryos", ylab = "Position changes before rescue time")

Subsetting data frame: to extract only values from the earliest stages where embryos got more trouble making a second hole to hatch

young_outcome<- subset(HC2016R, comst="2-3", select = NumPosChanADBefore2ndHole:Rescue)

str(young_outcome)

## 'data.frame':    167 obs. of  14 variables:
##  $ NumPosChanADBefore2ndHole       : int  3 7 1 1 17 1 1 14 16 1 ...
##  $ AccumADshakingR                 : int  NA NA NA NA NA NA NA NA NA NA ...
##  $ AccumshakingtimeRup2Fra         : int  NA 0 0 NA 0 NA NA NA 385 NA ...
##  $ AccumShakingR2sec               : num  NA 0 0 NA 0 ...
##  $ DisEndRup2Fr                    : int  NA NA NA NA NA NA NA NA NA NA ...
##  $ DisEndHatchingStartFr           : int  1407 4838 916 5182 6511 2222 213 8227 5455 187 ...
##  $ CompDurFr                       : int  NA NA 6 NA 7 NA NA NA NA NA ...
##  $ CompressionSec                  : num  NA NA 0.2 NA 0.23 NA NA NA NA NA ...
##  $ CompressionSeen                 : int  0 0 1 0 1 0 0 0 0 0 ...
##  $ EXITDURATIONFr                  : int  2442 1318 342 4 3995 157 96 NA NA 147 ...
##  $ EXIT.DURATION.StarttoFullyOutsec: num  81.4 43.933 11.4 0.133 133.167 ...
##  $ NumHoles                        : Factor w/ 2 levels "1","2": 1 1 1 1 1 1 1 1 2 1 ...
##  $ FirstSecondHoleExit             : chr  "First" "First" "First" "First" ...
##  $ Rescue                          : Factor w/ 2 levels "0","1": 1 1 1 1 1 1 1 1 1 1 ...

par(mfrow=c(1,2))
hist(young_outcome$NumPosChanADBefore2ndHole[young_outcome$Rescue=="1"], main = "Rescued embryos")
hist(young_outcome$NumPosChanADBefore2ndHole[young_outcome$Rescue=="0"], main = "Successfully hatched")

Boxplot Outcome (2-3)

boxplot(data = young_outcome, NumPosChanADBefore2ndHole ~ Rescue, col = c("burlywood2", 
    "lightpink1"), ylab =  "Position changes before rescue time")

wilcox.test(NumPosChanADBefore2ndHole~Rescue, data = young_outcome, correct=FALSE)

## 
##  Wilcoxon rank sum test
## 
## data:  NumPosChanADBefore2ndHole by Rescue
## W = 160, p-value = 8.265e-08
## alternative hypothesis: true location shift is not equal to 0

    *0= hatched successfully
    *1= rescued/unable to hatch
    *Orginal JMP: p-value = < 0.0001 (N= 10,14)
    *Replication: p-value = 8.265e-08 (< 0.0001)

Developmental stages (successfully hatched only)

boxplot(data = success_hat, NumPosChanADBefore2ndHole ~ comst, col = c("burlywood2", 
    "lightpink1", "purple"), ylab = "Position changes before 2nd hole", xlab = "grouped stages", main= "Successfully hatched embryos after displacement")

kruskal.test(NumPosChanADBefore2ndHole ~ comst,
             data =success_hat)

## 
##  Kruskal-Wallis rank sum test
## 
## data:  NumPosChanADBefore2ndHole by comst
## Kruskal-Wallis chi-squared = 2.7825, df = 2, p-value = 0.2488

    *Original: p-value = 0.16 (N = 14, 53, 84)
    *Replication: p-value = 0.2488
    

Exit from egg: fist or second rupture

boxplot(data = success_hat, NumPosChanADBefore2ndHole ~ NumHoles, col = c("burlywood2", 
    "lightpink1"), ylab = "Position changes before 2nd hole", xlab = "Exit from egg (1st or 2nd hole)", main= "Successfully hatched embryos after displacement")

wilcox.test(NumPosChanADBefore2ndHole~NumHoles, data = success_hat, correct=FALSE)

## 
##  Wilcoxon rank sum test
## 
## data:  NumPosChanADBefore2ndHole by NumHoles
## W = 3450.5, p-value = 3.33e-06
## alternative hypothesis: true location shift is not equal to 0

        * Original: p-value < 0.0001 (N = 41, 110)
        * Replication: p-value = 3.33e-06 (< 0.0001)
      

3.4 Body compression during exit: compare body compression observation and duration across different developmental stages

*Mosaic plot - compression seen across stages

totals1=table(success_hat$comst,success_hat$CompressionSeen)
mosaicplot(totals1, main = "Escape-hatching success after complications", xlab = "Embryonic developmental stage", ylab = "proportion of embryos", col=c("light green", "pink"))

mosaic(~ CompressionSeen + comst, data = success_hat, main = "Escape-hatching success after complications", shade= TRUE,legend=TRUE)

      * Original: p-value= 0.0032
      * Replication: p-value= 0.0036
      
      

• Boxplot compression duration in seconds

p <- ggplot(data=success_hat, aes(x=groupst, y= CompressionSec, fill=groupst))+stat_boxplot(geom='errorbar', linetype=1, width=0.5)+  #whiskers
geom_boxplot(outlier.shape=1)
p <- p + geom_boxplot()+ stat_summary(fun.y = mean, geom = "point", shape=5, size=5, fill="white")
p <- p + theme(axis.text.x=element_text(angle=0))
p <- p + ylab("Compression Duration (sec)") + xlab("Developmental stage")
p <- p + stat_n_text(y.pos = 70)
p <- p + theme_classic()+ scale_fill_manual(breaks = c("2,3", "4,7"), 
                       values=c("lightskyblue", "mediumaquamarine"))
p

## Warning: Removed 124 rows containing non-finite values (stat_boxplot).

## Warning: Removed 124 rows containing non-finite values (stat_boxplot).

## Warning: Removed 124 rows containing non-finite values (stat_boxplot).

## Warning: Removed 124 rows containing non-finite values (stat_summary).

## Warning: Removed 124 rows containing non-finite values (stat_n_text).

• Compression duration (s)

boxplot(CompressionSec ~ groupst, data=success_hat)

    * Original: p-value = 0.019 (N = 7, 21)
    * Replicate: p-value = 0.02087

wilcox.test(CompressionSec ~groupst, data=success_hat)

## Warning in wilcox.test.default(x = c(53.27, 7.5, 19.4, 15.4, 9.53, 0.03, :
## cannot compute exact p-value with ties

## 
##  Wilcoxon rank sum test with continuity correction
## 
## data:  CompressionSec by groupst
## W = 128, p-value = 0.02087
## alternative hypothesis: true location shift is not equal to 0

3.5. Exit duration: compare the speed of the hatching process (after displacement) across different developmental stages

• Boxplot exit duration in seconds

p <- ggplot(data=success_hat, aes(x=Stage, y=EXIT.DURATION.StarttoFullyOutsec, fill=Stage))+stat_boxplot(geom='errorbar', linetype=1, width=0.5)+  #whiskers
geom_boxplot(outlier.shape=1)
p <- p + geom_boxplot()+ stat_summary(fun.y = mean, geom = "point", shape=5, size=5, fill="white")
p <- p + theme(axis.text.x=element_text(angle=0))
p <- p + ylab("Exit duration (sec)")
p <- p + stat_n_text(y.pos = 170)
p <- p + theme_classic()+ scale_fill_manual(breaks = c("2", "3", "4", "5", "6", "7"), 
                       values=c("lightskyblue", "lightskyblue", "mediumaquamarine","mediumturquoise" ,"mediumturquoise","olivedrab2"))
p

## Warning: Removed 42 rows containing non-finite values (stat_boxplot).

## Warning: Removed 42 rows containing non-finite values (stat_boxplot).

## Warning: Removed 42 rows containing non-finite values (stat_boxplot).

## Warning: Removed 42 rows containing non-finite values (stat_summary).

## Warning: Removed 42 rows containing non-finite values (stat_n_text).

      *Original: p-value= 0.0006
      *Replication: p-value= 0.0002669

kruskal.test(EXIT.DURATION.StarttoFullyOutsec ~ Stage,
             data =success_hat)

## 
##  Kruskal-Wallis rank sum test
## 
## data:  EXIT.DURATION.StarttoFullyOutsec by Stage
## Kruskal-Wallis chi-squared = 23.533, df = 5, p-value = 0.0002669

4. Figure

5. Conclusions

• Embryos escaped either by re-finding their first hole or by making another. Few Stage-2 embryos made a second hole but the likelihood of doing so increased rapidly.

• Many Stage 2 embryos changed position many times, presumably in effort to re-find their first hole. More developed embryos also searched, but others immediately made another hole.

• Body compression during exit decreased with development; later stages made larger holes and passed through them more rapidly, likely due to both increased ennzyme reserves and greater muscular strength.

6. References

1.Salazar-Nicholls, M.J., K.D. Escobar & K.M. Warkentin. January 2017. Development of hatching ability in red-eyed treefrogs: escape from complications. Society for Integrative and Comparative Biology Meeting, New Orleans, LA.

2. Cohen, K. L., Seid, M. A., & Warkentin, K. M. (2016). How embryos escape from danger: the mechanism of rapid, plastic hatching in red-eyed treefrogs. Journal of Experimental Biology, 219(12), 1875-1883.

R Markdown

This is an R Markdown document. Markdown is a simple formatting syntax for authoring HTML, PDF, and MS Word documents. For more details on using R Markdown see http://rmarkdown.rstudio.com.

When you click the Knit button a document will be generated that includes both content as well as the output of any embedded R code chunks within the document. You can embed an R code chunk like this:

summary(cars)

##      speed           dist       
##  Min.   : 4.0   Min.   :  2.00  
##  1st Qu.:12.0   1st Qu.: 26.00  
##  Median :15.0   Median : 36.00  
##  Mean   :15.4   Mean   : 42.98  
##  3rd Qu.:19.0   3rd Qu.: 56.00  
##  Max.   :25.0   Max.   :120.00

Including Plots

You can also embed plots, for example:

Note that the echo = FALSE parameter was added to the code chunk to prevent printing of the R code that generated the plot.